Google Unveils AI Co-Scientist: A Revolutionary Tool for Scientific Discovery

Google's AI Co-Scientist : The Future of Scientific Breakthroughs?

Google has introduced an advanced AI system known as the "AI Co-Scientist," designed to reshape the way researchers approach scientific challenges. Acting as a virtual collaborator, this system uses innovative computational techniques to generate hypotheses, refine ideas, and propose innovative research directions. By enhancing research efficiency and streamlining the discovery process, the AI Co-Scientist is already demonstrating its ability to tackle some of the most complex problems in science and medicine. Its potential to transform the research landscape is evident, offering researchers a powerful tool to accelerate progress. The AI Co-Scientist is already proving its worth, diving into some of the most intricate problems in medicine and science with astonishing speed and precision. From identifying potential cancer treatments to uncovering mechanisms of bacterial resistance in record time, this multi-agent system is reshaping the research landscape. But how does it work, AI Grid explains more about what makes it so fantastic. Purpose-built for collaboration, scientists can interact with the system in many ways, including by directly providing their own seed ideas for exploration or by providing feedback on generated outputs in natural language. The AI Co-Scientist functions as a network of specialized agents, each tasked with addressing a specific aspect of the research process. These agents work collaboratively to generate, critique, and refine scientific hypotheses. The system employs an iterative self-critique mechanism, where ideas are continuously evaluated and improved to ensure optimal outcomes. This process is further enhanced by ranking and validation techniques, which help prioritize the most promising hypotheses for deeper exploration. Key features of the system include: These innovations allow the AI Co-Scientist to adapt and evolve, making it an increasingly effective tool for researchers. Its iterative approach ensures that the system remains aligned with the goals of scientific inquiry, providing robust support for complex research endeavors. The AI Co-Scientist excels in generating, evaluating, and refining research hypotheses. By simulating scientific debates and employing ranking tournaments, it can objectively assess the feasibility and potential impact of various ideas. One of its standout features is the use of ELO-based evaluation metrics, a ranking system commonly used in competitive environments, to measure the quality of its outputs with precision. The system's core strengths include: These capabilities position the AI Co-Scientist as an indispensable resource for researchers, particularly when addressing intricate scientific problems. Its ability to combine creativity with analytical rigor enables it to outperform traditional methods and, in some cases, even human experts. Here are more detailed guides and articles that you may find helpful on Google AI. The AI Co-Scientist has already demonstrated its utility across a variety of scientific disciplines, delivering impactful results in record time. Some notable applications include: These examples underscore the system's ability to deliver actionable insights with remarkable speed and accuracy. By addressing challenges that traditionally require extensive time and resources, the AI Co-Scientist is helping to advance scientific progress in unprecedented ways. The implications of the AI Co-Scientist extend far beyond its current applications. By significantly reducing research timelines, it has the potential to accelerate progress in addressing some of the world's most pressing challenges, including cancer, Alzheimer's disease, and other critical health issues. Its scalability ensures that it can be applied to a wide range of scientific and medical problems, making it a versatile tool for researchers across disciplines. As the system continues to evolve, future iterations are expected to enhance its capabilities even further. This could include improved methods for tackling longstanding scientific questions, allowing researchers to focus on the most promising ideas and experiments. By fostering collaboration between human expertise and artificial intelligence, the AI Co-Scientist is poised to redefine the boundaries of what is possible in scientific discovery. Google's AI Co-Scientist represents a significant advancement in the integration of artificial intelligence into scientific research. With its ability to generate, refine, and validate hypotheses at unprecedented speeds, this system is transforming how researchers approach complex problems. By accelerating the pace of discovery, it has the potential to unlock new possibilities in science and medicine, paving the way for breakthroughs that were once considered unattainable. As researchers continue to explore its capabilities, the AI Co-Scientist is set to play a pivotal role in shaping the future of innovation and discovery.

How Google's New AI Co-Scientist Tool Gives Us a Taste of Tomorrow's Workplace

In a post announcing the tool, Google explained how the new system would be used in a research setting. Essentially a scientist who has a specific topic to investigate -- like, say, discovering a new drug to tackle a particular disease -- would input that into the tool using natural language. The AI would then reply, much like any other chatbot, with a useful output -- in this case a hypothesis that the scientist can then test to either validate or invalidate their theory. The tool also does some of the work that goes into starting a new experiment by summarizing published literature about the topic, and suggesting an experimental approach. Google's blog post explains that the tool is actually a "multi-agent" system, tapping into what many think may be the next big thing in AI innovations. Using Gemini's ability to reason, synthesize data and perform long-term planning, the tool roughly models the actual intellectual process scientists use when tackling a novel problem -- the scientific method. In this case Google's system uses four AI agents called Generation, Reflection, Ranking and Evolution, refining its answers over and over in what Google calls a "self-improving cycle of increasingly high-quality and novel outputs." Essentially the tool cycles through lots of different ideas, checking how good they are and then spitting out what it thinks is the best output. Google is very careful to note that the tool is designed to be a scientific collaborator, to "help experts gather research and refine their work," and it's not meant to "automate the scientific process." What this means is that the AI co-scientist isn't designed to replace scientists, but instead may inspire researchers with novel ideas or otherwise speed up the process of investigating a thorny physics problem, or tackling a biological issue like antimicrobial resistance.

Google's AI 'Co-Scientist' Helps Unearth Research Ideas | PYMNTS.com

Scientific discovery is often a laborious, multi-step process that involves making observations, forming a hypothesis, conducting experiments, analyzing the data and drawing conclusions. One of the hardest steps in the process is coming up with a new idea to test. Now, Google has introduced an artificial intelligence (AI) system that can help scientists formulate new hypotheses, grounded in relevant research literature. Called "AI co-scientist," the AI system uses AI agents that generate an idea, debate the feasibility of the idea among themselves in a "tournament," and refine the suggested hypotheses before presenting it to the human scientist, according to a paper from Google, Stanford and other researchers. AI co-scientist has the "potential to accelerate scientists' efforts to address grand challenges in science and medicine," they wrote in a blog post. Importantly, the AI co-scientist is different from reasoning AI models, such as OpenAI's o3. It is also not the same as the "deep research" capabilities recently introduced by OpenAI, Google, Perplexity and others. AI co-scientist goes beyond them by producing new ideas, rather than summarizing existing ones. While the researchers focused on applying the AI co-scientist to biomedicine, they said it would work in any field. Also, they used Gemini 2.0 to power the AI system but it can be used with any large language model. The AI co-scientist is one of several areas where AI is being used to advance biomedicine. For example, Google DeepMind, whose researchers helped develop the AI co-scientist, solved a half-century challenge with AlphaFold 2, which predicted the 3D structure of all known proteins. AlphaFold 2 sped up drug discovery and vaccine development, leading to a Nobel Prize. While the AI co-scientist does a lot of heavy lifting, the researchers said it is not meant to replace the work of human scientists. It was built as a collaborative tool that still needs input and refinement from people. The use of AI in research papers has alsosparked concerns. The influx of AI-generated content in scientific literature is a growing problem, according to the Harvard Kennedy School's Shorenstein Center on Media, Politics and Public Policy. "The abundance of fabricated 'studies' seeping into all areas of the research infrastructure threatens to overwhelm the scholarly communication system and jeopardize the integrity of the scientific record," the institution wrote in September. Last week, the U.S. and U.K. offered fellowships to study the impact of AI in scientific research. The Alfred P. Sloan Foundation in the U.S. offered grants of up to $250,000 for up to two years for postdoctoral researchers, while the U.K. government set aside 4 million pounds ($5.1 million) for the effort. The AI co-scientist comprises several AI agents -- or bots that complete tasks given to them -- that interact with each other to do the work. Think of them as a scientist's own super-smart and fast research team of assistants. To start, the scientist prompts the AI co-scientist with a research goal or idea, along with preferences, experiment constraints and other attributes. This triggers the following actions: Source: Towards an AI co-scientist. To test the AI co-scientist, the team gave it several tasks, once of which was in the field of antibiotic resistance. For years, scientists have been trying to understand how bacteria becomes resistant to antibiotics, which is a big challenge in medicine. One way is through mobile genetic elements -- pieces of DNA that can move between different bacteria, spreading antibiotic resistance genes. The researchers focused on a special type of mobile genetic element called cf-PICIs (capsid-forming phage-inducible chromosomal islands). These can transfer between different bacterial species much more easily than typical viruses or other similar elements. It lets them spread traits, such as virulence and antibiotic resistance, across many bacteria. The researchers wanted to know how cf-PICIs are able to spread so widely among different bacteria. This knowledge could help scientists develop solutions to stop antibiotic resistance from spreading. The researchers' secret: Another study had already found a novel mechanism to explain how cf-PICIs spread, but it was not yet public. They wanted to see if the AI co-scientist could come up with the same answer. They tasked the AI co-scientist with generating a research proposal on this topic. They gave it only basic background information about cf-PICIs and two relevant research papers -- one explaining their discovery and another about using computers to study bacterial genetics. The result: Within two days, the AI co-scientist came up with this answer: cf-PICIs elements interact with diverse phage tails to expand their host range. "This finding was experimentally validated in the independent research study, which was unknown to the co-scientist during hypothesis generation," the authors wrote. While the AI co-scientist still needs to go through more testing, the team is hopeful. "These results ... demonstrate the potential to augment biomedical and scientific discovery and usher an era of AI-empowered scientists."

Google Goes Beyond Deep Research Tools, Introduces a New Co-Scientist System

The tool uses a multi-agent framework, powered by test time compute to aid scientists in research. Google today announced a new multi-agent AI system called 'AI co-Scientist' - a tool built with its Gemini 2.0 model. The system is designed to function as a 'virtual scientific collaborator' to help scientists generate novel research and hypotheses. While most big AI firms, such as OpenAI, xAI, and Perplexity, are betting big on a deep research agent, Google seems to have gone further. "Beyond standard literature review, summarisation and 'deep research' tools, the AI co-scientist system is intended to uncover new, original knowledge and to formulate demonstrably novel research hypotheses and proposals," said Google in a blog post. Google also says the tool uses a multi-agent system focusing on various research steps to automatically iterate, generate, evaluate, and refine hypotheses. Further, the tool uses web search, and 'specialised AI models' to enhance the quality of the generated hypotheses. "The AI co-scientist parses the assigned goal into a research plan configuration, managed by a Supervisor agent. The Supervisor agent assigns the specialised agents to the worker queue and allocates resources," read the blog post from Google. More importantly, the tool is also said to utilise test time computation to improve its outputs. Google also revealed that the tool outperforms several state of the art reasoning models. The tool was also tested through validation of the hypotheses it generated by laboratory experiments in drug repurposing, target discovery, and antimicrobial resistance. The AI co-scientist's successful validations include novel drug candidates for leukemia, potential anti-fibrotic targets for liver fibrosis, and insights into antimicrobial resistance mechanisms. Google has published a detailed technical paper outlining the AI co-scientist's capabilities, benchmark results, and real world performance. The tool can be accessed through Google's Trusted Tester Program. Interested research organisations can apply here. Similarly, American chipmaking giant AMD collaborated with Johns Hopkins University last month and introduced a new research study titled 'Agent Laboratory: and Using LLM Agents as Research Assistants'. The framework accepts a human-provided research idea. It will process the same across three stages, namely literature review, experimentation, and report writing, which includes a code repository and a research report. "By integrating specialised autonomous agents guided by human oversight, our approach can help researchers spend less time on repetitive tasks and more time on the creative, conceptual aspects of their work," said AMD researchers.

Google's New Multi-Agent AI Tool Can Collaborate With Scientists

Google unveiled a new artificial intelligence (AI) system on Wednesday that can assist scientists in the scientific discovery process. Dubbed AI co-scientist, the tool is powered by the Gemini 2.0 AI model and includes a multi-agent system that specialises in different tasks in the discovery process. The Mountain View-based tech giant stated that the AI system is designed to be a collaborative tool and not replace the human scientist. It is currently not available in the public domain, with Google planning to further test the system before releasing it. In a blog post, the tech giant detailed the new AI system. The multi-agent tool can mirror the reasoning process that is used by the scientific community for novel discovery. Some of the tasks it can perform include standard literature review, summarisation, researching hypotheses and proposals, as well as completing specific research objectives. Notably, the AI co-scientist also features the Deep Research tool recently released to Gemini Advanced users. The AI system features several specialised AI agents including Generation, Reflection, Ranking, Evolution, Proximity and Meta-review. Google says these agents were inspired by popular scientific methods. These agents can complete individual tasks as well as interact with each other to iteratively generate, evaluate, and refine hypotheses. These agents are also monitored by a supervisor agent. "We're seeing promising early results in important research areas like liver fibrosis treatments, antimicrobial resistance, and drug repurposing," said Google CEO Sundar Pichai, in a post on X (formerly known as Twitter). To start the process, scientists can interact with the system by specifying the research goal in natural language. They can also suggest their seed ideas and proposals for the AI to develop hypotheses around them. While the AI co-scientist works, the scientist can also provide feedback on the generated output to further fine-tune the process. The AI system uses tools such as web search and specialised AI models to enhance the grounding and quality of generated output. The details of how the tool works are detailed in a research paper. One of the unique aspects of this AI system is the test-time compute scaling capability that lets it second guess and verify its responses. This allows the AI co-scientist to iteratively reason, evolve, and improve outputs. However, it should be mentioned that Google's AI co-scientist is not capable of true innovation, which is often a prerequisite to making a scientific discovery. In the end, all of its information and hypotheses come from either its existing database or from web searches. While its reasoning capability does allow it to expand on ideas and test its validity, its role is unlikely to be larger than that of an assistant. The full capability of the AI system can only be assessed once the tool is released. Currently, Google is evaluating its strengths and limitations in science and biomedicine. The company has started a Trusted Tester Programme, through which it is enabling access to the AI co-scientist to research organisations. Those interested can join the programme by filling out this Google form.

Google's new AI generates hypotheses for researchers

Over the past few years, Google has embarked on a quest to jam generative AI into every product and initiative possible. Google has robots summarizing search results, interacting with your apps, and analyzing the data on your phone. And sometimes, the output of generative AI systems can be surprisingly good despite lacking any real knowledge. But can they do science? Google Research is now angling to turn AI into a scientist -- well, a "co-scientist." The company has a new multi-agent AI system based on Gemini 2.0 aimed at biomedical researchers that can supposedly point the way toward new hypotheses and areas of biomedical research. However, Google's AI co-scientist boils down to a fancy chatbot. A flesh-and-blood scientist using Google's co-scientist would input their research goals, ideas, and references to past research, allowing the robot to generate possible avenues of research. The AI co-scientist contains multiple interconnected models that churn through the input data and access Internet resources to refine the output. Inside the tool, the different agents challenge each other to create a "self-improving loop," which is similar to the new raft of reasoning AI models like Gemini Flash Thinking and OpenAI o3. This is still a generative AI system like Gemini, so it doesn't truly have any new ideas or knowledge. However, it can extrapolate from existing data to potentially make decent suggestions. At the end of the process, Google's AI co-scientist spits out research proposals and hypotheses. The human scientist can even talk with the robot about the proposals in a chatbot interface. You can think of the AI co-scientist as a highly technical form of brainstorming. The same way you can bounce party-planning ideas off a consumer AI model, scientists will be able to conceptualize new scientific research with an AI tuned specifically for that purpose. Today's popular AI systems have a well-known problem with accuracy. Generative AI always has something to say, even if the model doesn't have the right training data or model weights to be helpful, and fact-checking with more AI models can't work miracles. Leveraging its reasoning roots, the AI co-scientist conducts an internal evaluation to improve outputs, and Google says the self-evaluation ratings correlate to greater scientific accuracy. The internal metrics are one thing, but what do real scientists think? Google had human biomedical researchers evaluate the robot's proposals, and they reportedly rated the AI co-scientist higher than other, less specialized agentic AI systems. The experts also agreed the AI co-scientist's outputs showed greater potential for impact and novelty compared to standard AI models. This doesn't mean the AI's suggestions are all good. However, Google partnered with several universities to test some of the AI research proposals in the laboratory. For example, the AI suggested repurposing certain drugs for treating acute myeloid leukemia, and laboratory testing suggested it was a viable idea. Research at Stanford University also showed that the AI co-scientist's ideas about treatment for liver fibrosis were worthy of further study. This is compelling work, certainly, but calling this system a "co-scientist" is perhaps a bit grandiose. Despite the insistence from AI leaders that we're on the verge of creating living, thinking machines, AI isn't anywhere close to being able to do science on its own. That doesn't mean the AI-co-scientist won't be useful, though. Google's new AI could help humans interpret and contextualize expansive data sets and bodies of research, even if it can't understand or offer true insights.

Google builds AI 'co-scientist' tool to speed up research

Google has built an artificial intelligence laboratory assistant to help scientists accelerate biomedical research, as companies race to create specialised applications from the cutting-edge technology. The US tech group's so-called co-scientist tool helps researchers identify gaps in their knowledge and propose new ideas that could speed up scientific discovery. "What we're trying to do with our project is see whether technology like the AI co-scientist can give these researchers superpowers," said Alan Karthikesalingam, a senior staff clinician scientist at Google. Google's new tool comes as tech companies are spending billions of dollars on AI models and products, believing the technology can change industries from healthcare to energy and education. OpenAI, Perplexity and German drugmaker BioNTech and its London-based AI subsidiary InstaDeep have recently launched their own AI research tools, while Google DeepMind's AlphaFold has shown how the fast-developing technology can accelerate scientific research. Early tests of Google's new tool with experts from Stanford University, Imperial College London and Houston Methodist hospital found it was able to generate scientific hypotheses that showed promising results. The tool was able to reach the same conclusions -- for a novel gene transfer mechanism that helps scientists understand the spread of antimicrobial resistance -- as a new breakthrough from researchers at Imperial. Imperial's results were not in the public domain as they were being peer-reviewed in a top scientific journal. This showed that Google's co-scientist tool was able to reach the same hypothesis using AI reasoning in a matter of just days, compared with the years the university team spent researching the problem. The AI tool was also able to help researchers at Stanford find existing drugs that could be repurposed to treat liver fibrosis, a serious disease where scar tissue builds up in the organ. Google's co-scientist suggested two drug types that the Stanford scientists found helped with treating the illness. "We think it will be a tool that has the potential to change how we approach science," said José Penadés, a professor at Imperial's Department of Infectious Disease and the Fleming Initiative, who was part of the team behind the novel gene transfer mechanism study. The tool works by using several AI agents that mimic the scientific process. For example, one AI agent is specialised in generating ideas, and another in reflecting and reviewing those ideas, said Vivek Natarajan, research scientist at Google. The model is able to retrieve information from scientific papers and specialist databases that are freely available online, and other tools, such as AlphaFold. It then analyses the information it has been given and presents researchers with a ranked list of proposals with explanations and links to sources. Researchers can then refine these proposals. Tools such as Google's AI co-scientist could help scientists keep up with all the new information generated in their fields, said Jakob Foerster, an associate professor at the University of Oxford, who has also developed AI research tools. "I think it's super valuable," he said.

Google researchers develop AI co-scientist based on Gemini 2.0 - SiliconANGLE

Google LLC today introduced a new artificial intelligence tool designed to help researchers study scientific phenomena more efficiently. The company describes the tool as an AI co-scientist. It's powered by Gemini 2.0, the latest iteration of Google's flagship large language model family. The LLMs in the series can process multimodal data and have a tool use feature, which allows them to perform actions in external systems such as databases. Researchers interact with Google's AI co-scientist through a chatbot interface. The user specifies a goal, such as finding new clinical applications for an existing medicine, and the tool suggests potential ways of realizing that goal. The software generates a multi-paragraph research plan and finds academic papers with data relevant to the project. Users can customize the co-scientist's output in various ways. Instead of simply providing a research goal, a scientist could enter a proposal for reaching that goal and ask the AI to review the idea. Additionally, users can provide feedback on the co-scientist's initial prompt response to help it make improvements. "Beyond standard literature review, summarization and 'deep research' tools, the AI co-scientist system is intended to uncover new, original knowledge and to formulate demonstrably novel research hypotheses and proposals," Google researchers Juraj Gottweis and Vivek Natarajan wrote in a blog post. Under the hood, the AI co-scientist is powered by more than a half dozen AI agents. Those are machine learning programs that can perform actions with a high degree of autonomy. Each one carries out a different subset of the tasks involved in generating a research plan. The first agent, dubbed Generation, analyzes the scientific phenomenon the user wishes to study and generates several hypotheses that attempt to explain it. A second agent called Ranking then refines those hypotheses with the help of several auxiliary AI agents. One of the auxiliary agents, Proximity, removes duplicate research ideas. Another reviews the remaining research proposals with the help of publicly available scientific data. There's also a third agent, dubbed Evolution, that can simplify the AI co-scientist's output to make it easier to understand. The system uses an approach called test-time compute to generate hypotheses. The technique makes it possible to increase the quality of an AI model's output by boosting the amount of time and infrastructure it invests in generating prompt responses. Besides the Gemini 2.0 LLM series on which the co-scientist is based, test-time compute is also supported by several competing models including OpenAI's o1. The agents the co-scientist uses to perform research are coordinated by a supervisor agent. According to Google, one of its responsibilities is collecting statistics about the computations involved in processing a user prompt. Those statistics help the co-scientist determine when it should end processing and display its prompt response. To test the co-scientist's capabilities, Google asked a group of scientists to present the system with 15 research goals. The participants determined that the AI's responses "have higher potential for novelty and impact" than the output of competing models.

Google Launches AI Co-Scientist to Assist Researchers

Employs an Elo-based auto-evaluation metric for enhanced output quality. Google has launched an AI co-scientist, a new AI system powered by Gemini 2.0, designed to assist researchers in generating hypotheses, summarising scientific literature, and proposing experimental approaches. The system operates through a chatbot interface, where users specify a research goal, and the AI provides structured insights. It employs multiple AI agents to refine hypotheses, filter redundant ideas, and simplify research outputs. Currently, it is available to scientists in Google's Trusted Tester Program as an early access. Also Read: Perplexity Launches Deep Research for AI-Powered Expert Analysis "We introduce AI co-scientist, a multi-agent AI system built with Gemini 2.0 as a virtual scientific collaborator to help scientists generate novel hypotheses and research proposals, and to accelerate the clock speed of scientific and biomedical discoveries," Google said in a blog post on February 19. In a blog post, the Google Research team stated that unmet needs in the modern scientific discovery process, combined with recent AI advances, including the ability to synthesize across complex subjects and perform long-term planning and reasoning, have led to the development of AI co-scientist system. "Built on Gemini 2.0, AI co-scientist is designed to mirror the reasoning process underpinning the scientific method. Beyond standard literature review, summarization and 'deep research' tools, the AI co-scientist system is intended to uncover new, original knowledge and to formulate demonstrably novel research hypotheses and proposals, building upon prior evidence and tailored to specific research objectives." The AI co-scientist uses a coalition of specialized agents -- Generation, Reflection, Ranking, Evolution, Proximity and Meta-review -- which Google says are inspired by the scientific method itself. These agents use automated feedback to iteratively generate, evaluate, and refine hypotheses, resulting in a self-improving cycle of increasingly high-quality and novel outputs. Also Read: OpenAI Launches Deep Research: AI Agent for In-Depth Web Analysis Google stated that scientists can interact with the system in many ways, including directly providing their own seed ideas for exploration or offering feedback on generated outputs in natural language. The AI co-scientist also uses tools, like web-search and specialized AI models, to enhance the grounding and quality of generated hypotheses. The system's self-improvement cycle leverages an Elo-based auto-evaluation metric, which has been shown to correlate with higher-quality outputs. Expert assessments confirm that the AI co-scientist consistently outperforms state-of-the-art AI models and, in some cases, even human researchers. The AI co-scientist's potential has been demonstrated through real-world laboratory experiments, including: Drug Repurposing for Acute Myeloid Leukemia (AML): The system identified novel drug candidates, later validated in laboratory tests. "The AI co-scientist proposed novel repurposing candidates for acute myeloid leukemia (AML). Subsequent experiments validated these proposals, confirming that the suggested drugs inhibit tumor viability at clinically relevant concentrations in multiple AML cell lines," Google stated. Target Discovery for Liver Fibrosis: AI-suggested epigenetic targets showed promising anti-fibrotic effects in human liver organoids, with findings to be published by Stanford University researchers. According to Google, AI-assisted target discovery helps to streamline the process of experimental validation, potentially helping to reduce development time costs. Antimicrobial Resistance Research: The AI independently rediscovered a novel bacterial gene transfer mechanism, aligning with prior experimental findings from Imperial College London. Also Read: Google Expands Gemini 2.0 Lineup with New AI Models and Updates "The AI co-scientist represents a promising advance toward AI-assisted technologies for scientists to help accelerate discovery. Its ability to generate novel, testable hypotheses across diverse scientific and biomedical domains -- some already validated experimentally -- and its capacity for recursive self-improvement with increased compute, demonstrate its potential to accelerate scientists' efforts to address grand challenges in science and medicine," Google added. Despite its success, researchers acknowledge areas for improvement, including enhanced enhanced literature reviews, factuality checking, cross-checks with external tools, auto-evaluation techniques, and larger-scale evaluation. To refine the system further, a Trusted Tester Program is being launched, allowing research institutions to explore and contribute to the AI's development.

Can Google's new research assistant AI give scientists 'superpowers'?

Researchers who have been given access to Google's new AI "co-scientist" tool are enthusiastic about its potential, but it isn't yet clear whether it can make truly novel discoveries Google has unveiled an experimental artificial intelligence system that "uses advanced reasoning to help scientists synthesize vast amounts of literature, generate novel hypotheses, and suggest detailed research plans", according to its press release. "The idea with [the] 'AI co-scientist' is to give scientists superpowers," says Alan Karthikesalingam at Google. The tool, which doesn't have an official name yet, builds on Google's Gemini large language models. When a researcher asks a question or specifies a goal - to find a new drug, say - the tool comes up with initial ideas within 15 minutes. Several Gemini agents then "debate" these hypotheses with each other, ranking them and improving them over the following hours and days, says Vivek Natarajan at Google. During this process, the agents can search the scientific literature, access databases and use tools such as Google's AlphaFold system for predicting the structure of proteins. "They continuously refine ideas, they debate ideas, they critique ideas," says Natarajan. Google has already made the system available to a few research groups, which have released short papers describing their use of it. The teams that tried it are enthusiastic about its potential, and these examples suggest the AI co-scientist will be helpful for synthesising findings. However, it is debatable whether the examples support the claim that the AI can generate novel hypotheses. For instance, Google says one team used the system to find "new" ways of potentially treating liver fibrosis. However, the drugs proposed by the AI have previously been studied for this purpose. "The drugs identified are all well established to be antifibrotic," says Steven O'Reilly at UK biotech company Alcyomics. "There is nothing new here." While this potential use of the treatments isn't new, team member Gary Peltz at Stanford University School of Medicine in California says two out of three drugs selected by the AI co-scientist showed promise in tests on human liver organoids, whereas neither of the two he personally selected did - despite there being more evidence to support his choices. Peltz says Google gave him a small amount of funding to cover the costs of the tests. In another paper, José Penadés at Imperial College London and his colleagues describe how the co-scientist proposed a hypothesis matching an unpublished discovery. He and his team study mobile genetic elements - bits of DNA that can move between bacteria by various means. Some mobile genetic elements hijack bacteriophage viruses. These viruses consist of a shell containing DNA plus a tail that binds to specific bacteria and injects the DNA into it. So, if an element can get into the shell of a phage virus, it gets a free ride to another bacterium. One kind of mobile genetic element make its own shells. This type is particularly widespread, which puzzled Penadés and his team, because any one kind of phage virus can infect only a narrow range of bacteria. The answer, they recently discovered, is that these shells can hook up with the tails of different phages, allowing the mobile element to get into a wide range of bacteria. While that finding was still unpublished, the team asked the AI co-scientist to explain the puzzle - and its number one suggestion was stealing the tails of different phages. "We were shocked," says Penadés. "I sent an email to Google saying, you have access to my computer. Is that right? Because otherwise I can't believe what I'm reading here." However, the team did publish a paper in 2023 - which was fed to the system - about how this family of mobile genetic elements "steals bacteriophage tails to spread in nature". At the time, the researchers thought the elements were limited to acquiring tails from phages infecting the same cell. Only later did they discover the elements can pick up tails floating around outside cells, too. So one explanation for how the AI co-scientist came up with the right answer is that it missed the apparent limitation that stopped the humans getting it. What is clear is that it was fed everything it needed to find the answer, rather than coming up with an entirely new idea. "Everything was already published, but in different bits," says Penadés. "The system was able to put everything together." The team tried other AI systems already on the market, none of which came up with the answer, he says. In fact, some didn't manage it even when fed the paper describing the answer. "The system suggests things that you never thought about," says Penadés, who hasn't received any funding from Google. "I think it will be game-changing." Whether it really is game-changing will become clearer over time. Google's track record when it comes to claims about AI tools to help scientists is mixed. Its AlphaFold system is living up to the hype, winning the team behind it a Nobel prize last year. In 2023, however, the company announced that around 40 "new materials" had been synthesised with the help of its GNoME AI. Yet, according to a 2024 analysis by Robert Palgrave at University College London, not one of the synthesised materials was actually new. Despite his findings, Palgrave thinks AI can help scientists. "In general, I think AI has a huge amount to contribute to science if it is implemented in collaboration with experts in the respective fields," he says.

Google builds AI 'co-scientist' tool based on Gemini 2.0 for biomedical scientists. Here's what it can do

After quantum breakthrough Willow, Google has now built AI co-scientist tool based on Gemini 2.0 to help biomedical scientists. This new artificial intelligence tool has been designed to help researchers study scientific phenomena more efficiently. Researchers interact with Google's AI co-scientist through a chatbot interface and users can customize the co-scientist's output in various ways.Tech giant Google has developed an AI tool to act as a virtual collaborator for biomedical scientists, the US blue chip said on Wednesday. The new tool, tested by scientists at Stanford University in the US and Imperial College London, uses advanced reasoning to help scientists synthesize vast amounts of literature and generate novel hypotheses, the company said. AI is being increasingly deployed in the workplace, from answering calls to carrying out legal research, following the success of ChatGPT and similar models over the past year. "Beyond standard literature review, summarization and 'deep research' tools, the AI co-scientist system is intended to uncover new, original knowledge and to formulate demonstrably novel research hypotheses and proposals," Google researchers Juraj Gottweis and Vivek Natarajan wrote in a blog post. ALSO READ: After Google's Willow, Microsoft unveils new quantum chip 'Majorana 1'- powered by new state of matter: 10 points -Google in its blog post said it has launched an AI co-scientist, a new AI system built on Gemini 2.0 designed to aid scientists in creating novel hypotheses and research plans. -Researchers can specify a research goal -- for example, to better understand the spread of a disease-causing microbe -- using natural language, and the AI co-scientist will propose testable hypotheses, along with a summary of relevant published literature and a possible experimental approach, the tech giant said. -AI co-scientist is a collaborative tool to help experts gather research and refine their work -- it's not meant to automate the scientific process. We're excited to see how researchers will use the system for their research, Google said. -In an experiment on liver fibrosis, Google reported that all approaches suggested by the AI co-scientist demonstrated promising activity and potential to inhibit disease causes. Google noted that the tool showed the capacity to enhance solutions generated by experts over time. ALSO READ: Amid divorce rumours with Bianca Censori after Grammys shocker, Kanye West re-unites with ex Kim Kardashian -"While this is a preliminary finding requiring further validation, it suggests a promising avenue for capable AI systems... to augment and accelerate the work of expert scientists," the company stated. -The AI co-scientist system is reportedly not intended to fully automate the scientific process. Instead, it is designed for collaboration, allowing experts to interact with the tool using simple natural language and provide feedback, including their own hypotheses for experimental testing. -Professor José Penadés, from Imperial's Department of Infectious Disease and the Fleming Initiative (a partnership between Imperial College London and Imperial College Healthcare NHS Trust) who co-led the experimental work, told The Verdict: "When the Google research team approached us to test its AI platform, we realised we needed to task it with the same scientific questions that we had already explored ourselves and used as the basis of our experimental work." ALSO READ: Will Americans get a refund of $5000 from Elon Musk? All about 'DOGE dividend' plan -"This effectively meant that the algorithm was able to look at the available evidence, analyse the possibilities, ask questions, design experiments, and propose the very same hypothesis that we arrived at through years of painstaking scientific research, but in a fraction of the time." -Scientists involved in the project emphasised that the tool is meant to complement, not replace, researchers. "We expect that it will... increase, rather than decrease scientific collaboration," Google scientist Vivek Natarajan said. -Earlier in February 2025, Google launched a new class of AI models within its Gemini family, providing a cost-effective alternative to models from competitors, including low-cost options from Chinese company DeepSeek. (With inputs from Reuters)

Google develops AI co-scientist to aid researchers

LONDON (Reuters) - Google has developed an AI tool to act as a virtual collaborator for biomedical scientists, the U.S. blue chip said on Wednesday. The new tool, tested by scientists at Stanford University in the U.S. and Imperial College London, uses advanced reasoning to help scientists synthesize vast amounts of literature and generate novel hypotheses, the company said. AI is being increasingly deployed in the workplace, from answering calls to carrying out legal research, following the success of ChatGPT and similar models over the past year. Google's AI unit, DeepMind, has made science a priority, and DeepMind boss Demis Hassabis was a co-recipient of a Nobel Prize in Chemistry last year for technology developed in the AI unit. In an experiment on liver fibrosis, Google said all the approaches suggested by its new AI co-scientist showed promising activity and potential to inhibit causes of disease. It showed the capacity to improve solutions generated by experts over time, Google added. "While this is a preliminary finding requiring further validation, it suggests a promising avenue for capable AI systems... to augment and accelerate the work of expert scientists," it said. The scientists who worked on the project said it would complement rather than replace researchers. "We expect that it will... increase, rather than decrease scientific collaboration," Google scientist Vivek Natarajan said. (Reporting by Muvija M; Additional reporting by Kenrick Cai in San Francisco; Editing by Jan Harvey)

We're launching a new AI system for scientists.

Today Google is launching an AI co-scientist, a new AI system built on Gemini 2.0 designed to aid scientists in creating novel hypotheses and research plans. Researchers can specify a research goal -- for example, to better understand the spread of a disease-causing microbe -- using natural language, and the AI co-scientist will propose testable hypotheses, along with a summary of relevant published literature and a possible experimental approach. AI co-scientist is a collaborative tool to help experts gather research and refine their work -- it's not meant to automate the scientific process. We're excited to see how researchers will use the system for their research. Scientists who are part of our Trusted Tester Program will have early access to AI co-scientist -- you can learn more about how this new tool works on the Google Research blog.

Google's AI Co-scientist is 'test-time scaling' on steroids. What that means for research

Google on Wednesday said it has tweaked its Gemini 2.0 large language model artificial intelligence offering to make it generate novel scientific hypotheses in a fraction of the time taken by teams of human lab researchers. The company bills the "AI Co-scientist" version of Gemini as "a promising advance toward AI-assisted technologies for scientists to help accelerate discovery," and a program meant to be run with a human "in the loop" to "act as a helpful assistant and collaborator to scientists and to help accelerate the scientific discovery process." Also: Google's Gemini 2.0 AI promises to be faster and smarter via agentic advances It's also a demonstration of how so-called reasoning AI models are now driving the use of computing resources higher and higher, to cross-reference, evaluate, rank, sort, sift, and do lots of other things -- all after the prompt has been typed by the user. In an audacious mash-up of scientific publishing and marketing, Google's researchers published a technical paper describing a hypothesis generated by Co-scientist simultaneously with a paper published by a group of human scientists at Imperial College London, with the same hypothesis. The Co-scientist hypothesis, concerning a specific fashion in which bacteria evolve to form new pathogens, took two days to produce, whereas the human-produced work was the result of a decade of study and lab work, claims Google. Google describes the machine as a hypothesis-formulation machine that uses multiple agents. Given a scientist's research goal that has been specified in natural language, the AI Co-scientist is designed to generate novel research hypotheses, a detailed research overview, and experimental protocols. To do so, it uses a coalition of specialized agents: Generation, Reflection, Ranking, Evolution, Proximity, and Meta-review. The Co-scientist starts to work after the scientist types at the prompt their research goal "along with preferences, experiment constraints, and other attributes." Google insists the program goes beyond mere literature review to instead "uncover new, original knowledge and to formulate demonstrably novel research hypotheses and proposals, building upon prior evidence and tailored to specific research objectives." The modification of Gemini 2.0 emphasizes the use of "test-time scaling," where AI agents use increasing amounts of computing power to iteratively review and re-formulate their output. Test-time scaling has been seen most dramatically not only in Gemini, but also OpenAI's o1 model, and DeepSeek AI, all examples of so-called reasoning models that spend much more time responding to a prompt, generating intermediate results. The AI Co-scientist is a bit of test-time scaling on steroids. Also: What is Gemini? Everything you should know about Google's new AI model In the formal paper, authored by Juraj Gottweis of Google, and posted on the arXiv pre-print server, the authors specifically relate their work as a kind of enhancement of what DeepSeek's R1 model has pioneered: "Recent advancements, like the DeepSeek-R1 model, further demonstrate the potential of test-time compute by leveraging reinforcement learning to refine the model's "chain-of-thought" and enhance complex reasoning abilities over longer horizons. In this work, we propose a significant scaling of the test-time compute paradigm using inductive biases derived from the scientific method to design a multi-agent framework for scientific reasoning and hypothesis generation without any additional learning techniques." The Co-scientist is built from a selection of AI agents that can access external resources, relate Gottweis and team. "They are also equipped to interact with external tools, such as web search engines and specialized AI models, through application programming interfaces," they write. Also: What is sparsity? DeepSeek AI's secret, revealed by Apple researchers Where test-time scaling comes most into play is the notion of a "tournament," where the Co-scientist compares and ranks the multiple hypotheses it has generated. It does so using "Elo" scores, a common measurement system used to rank chess players and athletes. As Gottweis and team describe it, one of the agents, a "Ranking Agent," has the main responsibility of rating the differing hypotheses in a kind of competitive fashion: An important abstraction in the Co-scientist system is the notion of a tournament where different research proposals are evaluated and ranked, enabling iterative improvements. The Ranking agent employs and orchestrates an Elo-based tournament to assess and prioritize the generated hypotheses at any given time. This involves pairwise comparisons, facilitated by simulated scientific debates, which allow for a nuanced evaluation of the relative merits of each proposal. Also: What is DeepSeek AI? Is it safe? Here's everything you need to know The ranking is supposed to make the better hypotheses bubble up to the top. "This ranking serves to communicate to scientists an ordered list of research hypotheses and proposals aligned with the research goal," as they put it. Google claims the data show that more and more compute, and ranking and re-ranking, makes the hypotheses increasingly better as rated by human observers. According to fifteen human experts who reviewed the Co-scientist's output, the program gets better as it spends more computing time formulating hypotheses and evaluating them. "As the system spends more time reasoning and improving, the self-rated quality of results improves and surpasses models and unassisted human experts," the paper notes. The human observers generally gave Co-scientist "higher potential for novelty and impact, and preferred its outputs compared to other models," such as the unaltered Gemini 2.0 and OpenAI's o1 reasoning model. Given the emphasis on scaling computing effort, it's unfortunate that Gottweis and team nowhere in their 70-page technical report mention just how much computing was used for AI Co-scientist. The hypothesis, however, that they share, is that the rapid reduction in the cost of computing of the kind DeepSeek R1 demonstrates should make something like the Co-scientist usable by research labs broadly speaking. "The trends with distillation and inference time compute costs indicate that such intelligent and general AI systems are rapidly becoming more affordable and available," they note.